BSc Hons Data Science and Artificial Intelligence
Machine Learning Engineer with applied experience in industry and open-source ecosystems. At Amazon Web Services and Huawei Ireland Research Centre, engineered performant Rust and Python tooling for scalable ML workflows. Expertise spans concurrency, ETL systems, and cross-language integrations. Contributor to the Rust toolchain and academic ML research; motivated about building high-performance, interoperable solutions at scale.
Education
Experience
- Continuing previous open source development internally on AWS Kani and SageMaker to further support general infrastructure and tooling upgrades via Rust scripting and Python automation across existing data pipelines
- Working as part of Nemo (CloudWatch hot data store) optimizing large-scale real-time batching and streaming processes using Spark/SIMD with a strong focus on ingestion towards the R&D for cloudwatch agents
- Contributing to internal codebase through Git workflows, issue tracking, and collaborative reviews
- Benchmarking and prototyping data processing utilities, identifying bottlenecks in ingestion latency
- Contributed to Huawei’s Advanced Language Engineering lab’s joint research with Peking University to advance Rust for memory-safe, highly performant and concrete systems with a focus on furthering reach of current tooling for future LLM integration
- Optimized Rust language/tooling workflows (concurrency, I/O, and build-time ergonomics) and evaluated advanced Rust features for internal adoption
- Collaborated with the Rust Foundation community and upstream ecosystem; work focused on optimizing the Rust language and toolchain for trusted, production systems
- Produced prototypes and internal benchmarks that informed ADA Lab’s Rust adoption roadmap and trustworthy programming practices
- Engineered performant Rust based internal tools and infrastructure supporting ML and deep learning workflows for proprietary Pangu models, enhancing concurrency and memory safety
- Architected scalable ETL pipelines using Polars, Apache Arrow, and Parquet, processing 10M+ records daily
- Implemented Rust–Python bindings via PyO3 and maturin, reducing runtime for batch jobs by 40%
- Operated in a high-performance computing environment alongside leading global researchers
- Maintainer for rustup, the official Rust toolchain installer
- Introduced concurrent download strategies, improving install speed by 50%
- Refactored locking and synchronization primitives, reducing contention under load
- Facilitated RFCs and collaborated with Rust core contributors through PRs and issue triage